Dynamic Modeling and Optimization of a UV Water Disinfection System Using PSO for Enhanced Bacterial Reduction

Abstract

This paper presents a hybrid framework combining dynamic system modeling and Particle Swarm Optimization (PSO) to improve UV water disinfection. The reactor is modeled as a MIMO system, with flow rate (0.2–0.8 L/s) and UV intensity (5, 7, and 10 mW/cm<sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup>) as inputs, and UV dose and bacterial reduction as outputs. A dynamic transfer function is identified from experimental data, and its parameters are optimized via PSO to minimize the difference between simulated and measured results. Validation at 10 mW/cm<sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup> shows high predictive accuracy, and further simulations confirm the model’s robustness in capturing the system’s nonlinear behavior. Combining dynamic modeling with PSO provides a robust predictive tool for UV systems. This approach offers a reliable and efficient method to optimize UV disinfection under varying operating conditions.